Synthetic Pathway for the Production of Olivetolic Acid in Escherichia

Jul 5, 2018 - (19) Given the need for hexanoyl-CoA and malonyl-CoA in OLA ..... flow rate = 0.25 mL min–1; 0–2.5 min, 95% A and 5% B; 2.5–20 min...
0 downloads 0 Views 2MB Size
Article

Subscriber access provided by BOSTON COLLEGE

Synthetic pathway for the production of olivetolic acid in Escherichia coli Zaigao Tan, James Clomburg, and Ramon Gonzalez ACS Synth. Biol., Just Accepted Manuscript • DOI: 10.1021/acssynbio.8b00075 • Publication Date (Web): 05 Jul 2018 Downloaded from http://pubs.acs.org on July 7, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

1

Synthetic Pathway for the Production of Olivetolic Acid in Escherichia coli

2 Zaigao Tan1, James M Clomburg1, Ramon Gonzalez1,2*

3 4 1

5

Department of Chemical and Biomolecular Engineering, Rice University, Houston, TX, USA

6 2

7

Department of Bioengineering, Rice University, Houston, TX, USA

8 9

*

To whom correspondence should be addressed:

10

Department of Chemical and Biomolecular Engineering/Bioengineering

11

Rice University, 6100 Main Street, MS-667,

12

Houston, TX 77005, USA.

13

Phone: (713) 348-4893, Fax: (713) 348-5478

14

Email: [email protected]

15 16 17 18 19 20 21 22 23

1 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

24

ABSTRACT

25

Type III polyketide synthases (PKS IIIs) contribute to the synthesis of many economically

26

important natural products, most of which are currently produced by direct extraction from plants

27

or through chemical synthesis. Olivetolic acid (OLA) is a plant secondary metabolite sourced

28

from PKS III catalysis, which along with its prenylated derivatives has various pharmacological

29

activities. To demonstrate the potential for microbial cell factories to circumvent limitations of

30

plant extraction or chemical synthesis for OLA, here we utilize a synthetic approach to engineer

31

Escherichia coli for the production of OLA. In vitro characterization of polyketide synthase and

32

cyclase enzymes, OLA synthase and OLA cyclase, respectively, validated their requirement as

33

enzymatic components of the OLA pathway and confirmed the ability for these eukaryotic

34

enzymes to be functionally expressed in E. coli. This served as a platform for the combinatorial

35

expression of these enzymes with auxiliary enzymes aimed at increasing the supply of

36

hexanoyl-CoA and malonyl-CoA as starting and extender units, respectively. Through combining

37

OLA synthase and OLA cyclase expression with the required modules of a β-oxidation reversal

38

for hexanoyl-CoA generation, we demonstrate the in vivo synthesis of olivetolic acid from a

39

single carbon source. The integration of additional auxiliary enzymes to increase hexanoyl-CoA

40

and malonyl-CoA, along with evaluation of varying fermentation conditions enabled the

41

synthesis of 80 mg/L OLA. This is the first report of OLA production in E. coli, adding a new

42

example to the repertoire of valuable compounds synthesized in this industrial workhorse.

43

KEYWORDS Type III polyketide synthases (PKS III); Olivetolic acid (OLA); Synthetic

44

Biology; Natural Products.

2 ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

45 46

A broad diversity of natural products can be synthesized by type III polyketide synthases

47

(PKS IIIs).1, 2 Many of these products have been found to benefit human health, with PKS III

48

products and derivatives garnering significant research interest in recent years.1, 2 For instance,

49

anthocyanins, the water-soluble pigments from mulberry fruits, have been reported to be useful

50

in treating obesity, inflammation and cancer.3 Hyperforin, which is one of the primary active

51

constituents from extracts of Hypericum perforatum, can be used for the treatment of

52

depression.1 The monoaromatic compound olivetolic acid (OLA), a member of the PKS III

53

product class, holds promise for its pharmacological properties such as antimicrobial, cytotoxic,

54

and photoprotective activities.2, 4 In addition, OLA is a central intermediate in the synthesis of an

55

important class of pharmacological compounds, as it serves as the alkylresorcinol moiety during

56

the biosynthesis of cannabinoids, a class of products that are becoming increasingly important

57

due to their numerous pharmacological properties.5-7

58

Currently, the production of OLA and its derivatives is primarily through direct extraction

59

from plants 4, 8, 9 However, given that plants grow slowly and require at least several months for

60

the accumulation of these compounds, direct extraction suffers from long cycles.10 While plant

61

biotechnology offers the opportunity to improve natural product synthesis in native species, it is

62

difficult to precisely control the expression level of transgenes in plants and adapt to

63

industrial-scale production.11 While the chemical synthesis of OLA is another alternative that has

64

recently been reported,12 the structural complexity of most natural products dictates inherent

65

inefficiencies with total chemical synthesis, which suffers from low yield and high energy

66

waste.13 In contrast to these approaches, construction of microbial cell factories for production of

67

these value-added plant natural products is a promising strategy.13-16

3 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

68

Despite the structural complexity of the end product, the starting and extending units for

69

polyketide biosynthesis are often tractable acyl-coenzyme A (CoA) intermediates. In the case of

70

OLA biosynthesis, 3 iterations of 2-carbon additions (via decarboxylative condensation with

71

malonyl-CoA as the donor) to an initial hexanoyl-CoA primer results in the formation of

72

3,5,7-trioxododecanoyl-CoA, which can be subsequently cyclized to form OLA (Fig. 1). While it

73

was initially thought that the polyketide synthase (OLS) from C. sativa was solely responsible

74

for OLA biosynthesis, recombinant OLS was found to only synthesize olivetol, the

75

decarboxylated form of OLA.17 It has since been shown that olivetolic acid biosynthesis requires

76

a polyketide cyclase, i.e. OLA cyclase (OAC), in addition to OLS, which catalyzes a C2-C7

77

intramolecular aldol condensation of the 3,5,7-trioxododecanoyl-CoA intermediate with

78

carboxylate retention.18 Expression of OLS and OAC in Saccharomyces cerevisiae, along with

79

feeding of sodium hexanote, enabled the synthesis of 0.48 mg/L olivetolic acid in a 4 day

80

fermentation.18 This represents a promising first step toward the development of microbial cell

81

factories for the production of OLA that can be built upon to improve product synthesis from

82

biorenewable feedstocks.

83

A

potential

bottleneck

in

improving

product

synthesis

in

S.

cerevisiae

is

84

compartmentalization of acetyl-CoA metabolism, which results in the requirement for significant

85

engineering efforts for the production of acetyl-CoA-derived products.19 Given the need for

86

hexanoyl-CoA and malonyl-CoA in OLA synthesis, which are both commonly derived from

87

acetyl-CoA, here we explored the possibility of engineering Escherichia coli for OLA production.

88

In addition to its well-known physiology, metabolic network, and the ease of genetic

89

manipulation, E. coli has been engineered to produce a wide range of products from the

90

acetyl-CoA node, including those derived directly from malonyl-CoA.19-22 We utilize a

4 ACS Paragon Plus Environment

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

91

bottom-up, synthetic biology approach to develop a pathway for OLA production through testing

92

and validating the required PKS and cyclase components in addition to auxiliary enzymes for

93

generating the required precursors. Through the combinatorial expression of these enzymatic

94

components with the required modules of a β-oxidation reversal pathway to supply

95

hexanoyl-CoA23, 24, we demonstrate a functional biological pathway for the synthesis of OLA

96

from a single carbon source and further identified precursor supply as a major limiting factor for

97

product synthesis. The use of auxiliary enzymatic components aimed at increasing

98

hexanoyl-CoA and malonyl-CoA and systematic metabolic engineering efforts enabled the

99

synthesis of OLA at a titer of 80 mg/L, further demonstrating the viability of developing

100

microbial cell factories for the synthesis of plant-based natural products.

101 102

RESULTS AND DISCUSSION

103

Recruiting OLS and OAC for olivetolic acid production. A synthetic pathway for olivetolic

104

acid (OLA) biosynthesis in E. coli requires at least two catalytic enzymes, OLA synthase (OLS)

105

and OLA cyclase (OAC).17,

106

trichomes that catalyzes the formation of 3,5,7-trioxododecanoyl-CoA from a hexanoyl-CoA

107

primer and 3 malonyl-CoA extender units via decarboxylative Claisen condensation.17 This

108

3,5,7-trioxododecanoyl-CoA intermediate can then be cyclized by OAC via C2–C7

109

intramolecular aldol condensation to form OLA.18 In addition to the desired product, evidence

110

suggests that pathway byproducts, e.g. pentyl diacetic acid lactone (PDAL), hexanoyl triacetic

111

acid lactone (HTAL) and olivetol can also be formed through hydrolysis of intermediate

112

polyketide CoAs or spontaneous cyclization (Fig. 1A).18

113

18

OLS is a type III PKS (tetraketide synthase) from Cannabis

To confirm the ability for OLS and OAC expressed in E. coli to synthesize OLA from

5 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

114

hexanoyl-CoA and malonyl-CoA, in addition to evaluating potential by-products, we conducted

115

in vitro analysis of these enzymatic components (Fig. 1B). Codon optimized, His-tagged OLS

116

and OAC were expressed and purified from E. coli and utilized to determine product formation

117

in a reaction system including hexanonyl-CoA (primer) and malonyl-CoA (extender unit). As

118

seen in Figure 1B, incubation of OLS and OAC in the presence of these substrates resulted in

119

OLA synthesis. Pathway byproducts PDAL and olivetol were also detected in samples with OLS

120

only or including both OLS and OAC (Fig. 1B). These by-products were the only products

121

formed in the absence of OAC (i.e. assays with OLS only) confirming the indispensable nature

122

of the OAC component for OLA formation.18

123 124 125 126 127 128 129 130 131 132 133 134 135

Figure 1. Production of olivetolic acid (OLA) by recruiting OLS and OAC. (A) Biosynthetic pathway of OLA from hexanoyl-CoA. 3 malonyl-CoA extender units are added to the hexanoyl-CoA primer to form OLA through the Claisen condensation catalyzed by OLS and C2-C7 aldol cyclization catalyzed by OAC. Potential pathway byproducts, e.g. PDAL, HTAL and olivetol, can also be formed through hydrolysis of intermediate CoAs or spontaneous cyclization without carboxyl group retention. (B) In vitro production of OLA using recombinant and purified OLS and OAC core enzymes. Detailed MS identification of these OLA and by-products can be seen in Fig. S2. (C) In vivo production of OLA from resting cells biotransformations with E. coli BL21 (DE3). E. coli cells with the induced OLS and OAC from LB medium were collected and resuspended in fresh M9Y medium+2% (wt/v) glucose with 4 mM hexanoate, and cultured at 22 °C for 48 h. Blank, BL21 (DE3) with pETDuet-1 empty vector. PDAL, pentyl diacetic acid lactone, HTAL, hexanoyl triacetic acid lactone.

136

We next evaluated the ability to produce OLA in vivo through the construction of plasmid 6 ACS Paragon Plus Environment

Page 6 of 30

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

137

pET-P1-OLS-P2-OAC expressing codon-optimized versions of OLS and OAC. This plasmid was

138

transformed into E. coli BL21 (DE3), both the OLS and OAC have soluble expression (Fig. S1),

139

and the resulting strain (BL-OLS-OAC) enabled the production of OLA following 48 h

140

cultivation in biotransformation media with 4 mM hexanote (Fig. 1C). In addition to OLA, small

141

amounts of olivetol were also detected, implying that while heterologous OAC cyclized the

142

3,5,7-trioxododecanoyl-CoA intermediate into OLA, the potential for by-product formation is

143

also a concern in vivo. It should be noted that even with the small amounts of OLA produced (~

144

0.1 mg/L), this product was present in the fermentation broth (supernatant) instead of cell pellet

145

(Fig. 1C). The presence of OLA in supernatant might be due to export of OLA by a native E. coli

146

transporter or from cell lysis.

147

Impact of precursor supply on olivetolic acid production in vivo. While the above results

148

demonstrate the function of the required PKS and cyclase components in vivo, the low OLA

149

titers (~ 0.1 mg/L) (Fig. 2) require additional assessment of the overall limitations for product

150

synthesis. Given the functional expression and purification of OLS and OAC from E. coli, we

151

reasoned a major limitation for OLA production is the availability of required precursors,

152

opposed to issues with the expression or activity of these enzymes in E. coli. To determine the

153

potential to improve product synthesis by increasing precursor supply, we expanded our

154

synthetic approach through combinatorially expressing auxiliary enzymes for malonyl-CoA

155

and/or hexanoyl-CoA generation with the OLS and OAC components.

156

For increasing malonyl-CoA supply, two classes of enzymes for biosynthesis of

157

malonyl-CoA were evaluated. The first, malonyl-CoA synthetase (MCS) (EC 6.2.1.14), catalyzes

158

the formation of malonyl-CoA from malonate, CoA and ATP.26,

159

Bradyrhizobium japonicum27 was codon optimized and expressed in conjunction with OLS and

7 ACS Paragon Plus Environment

27

The MCS from

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

160

OAC in E. coli supplied with 12 mM sodium malonate. Consistent with our hypothesis,

161

increasing malonyl-CoA supply using this approach resulted in increased OLA titer, from 0.1

162

mg/L to 0.65 mg/L (p-value < 0.05) (Fig. 2). While this shows the importance of increasing

163

malonyl-CoA supply, MCS requires the addition of exogenous malonate. To generate increased

164

malonyl-CoA without malonate supplementation, we evaluated the overexpression of acetyl-CoA

165

carboxylase (ACC) (EC 6.4.1.2) catalyzing the carboxylation of acetyl-CoA in the presence of

166

ATP and bicarbonate (HCO3-). In E. coli, ACC consists of four different subunits, e.g. AccA,

167

AccB, AccC and AccD. Genes encoding the AccABCD complex were overexpressed (+ACC) in

168

the BL-OLS-OAC strain (resulting in BL-OLS-OAC-ACC). However, the engineered +ACC

169

strain did not improve OLA production in the absence of malonate (Fig. 2). ACC requires

170

acetyl-CoA as catalytic substrate, which is one of the most important central metabolites in E.

171

coli, participating in the TCA cycle, glyoxylate cycle, amino acid metabolism, and other

172

important pathways.28 Although individual +ACC overexpression could channel more

173

acetyl-CoA into malonyl-CoA available for OLA production, flux into other biosynthetic

174

pathways and thus cellular growth may be impaired. Consistently, we observed lower growth

175

with BL-OLS-OAC-ACC during the biotransformation (the highest OD550 was only 1.8)

176

compared with BL-OLS-OAC (the highest OD550 was 3.8). Impaired growth caused by ACC

177

overexpression has been observed in prior studies, in both E. coli 29 and S. cerevisiae. 30

8 ACS Paragon Plus Environment

Page 8 of 30

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193

ACS Synthetic Biology

Figure 2. Impact of auxiliary enzymes for increasing hexanoyl-CoA and malonyl-CoA supply on olivetolic acid (OLA) production in BL21 (DE3). Left, auxiliary enzymes employed for increasing hexanoyl-CoA (FadD/FadK) and malonyl-CoA (MCS/ACC). FadD and FadK were employed to form hexanoyl-CoA from hexanoate and CoA. MCS and ACC can form the malonyl-CoA extender unit through two different mechanisms. Middle, enzyme organization of OLS/OAC core enzymes and FadD/FadK, MCS/ACC auxiliary enzymes. The vector expressing OLS and OAC was constructed using multiple cloning site 1 (Mcs1) and multiple cloning site 2 (Mcs2) of the pETduet-1 plasmid respectively. FadD/FadK were constructed at the Mcs1 of pCDFduet-1 and MCS/ACC were constructed at the Mcs2 of pCDFduet-1. Right, OLA production using different combinations of auxiliary enzymes with OLS and OAC expression. E. coli cells expressing the indicated enzymes grown in LB medium were collected and resuspended in fresh M9Y medium with 2% (wt/v) glucose and 4 mM hexanoate. For strains harboring MCS, 12 mM malonate sodium was also included. Values represent the average of at least three biological replicates with error bars indicating standard deviation. FadD, long chain fatty acyl-CoA synthetase or ligase; FadK, short chain fatty acyl-CoA synthetase; MCS, malonyl-CoA synthetase; ACC, acetyl-CoA carboxylase.

194

While these initial experiments were conducted in the presence of hexanoate, the

195

conversion of hexanoate to hexanoyl-CoA in these strains may be inefficient due to potential low

196

expression levels of E. coli native fatty acyl-CoA synthetase(s). To evaluate the impact of the

197

hexanoyl-CoA pool on OLA production, two different E. coli native fatty acyl-CoA synthetases

198

were overexpressed individually. FadK has been reported as an acyl-CoA synthetase which is

199

primarily active on acetylation of short chain fatty acids (C6-C8).31, 32 However, we found that

200

overexpression of FadK had no impact on OLA production under these conditions (Fig. 2). We

201

also explored another native E. coli fatty acyl-CoA synthetase, FadD, which has broad chain

202

length specificity, with maximal activities associated with fatty acids ranging in length from C12

9 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

203

to C18.33, 34 Prior studies have showed that, compared to FadK, despite the lower specificity for

204

C6-C8, FadD exhibits a higher catalytic activity on C6-C8 fatty acids.32 The overexpression of

205

FadD in combination with OLS and OAC (strain BL-OLS-OAC-FadD) resulted in slight

206

increases in OLA titer (Fig. 2). Combining this hexanoyl-CoA generating module with MCS, the

207

highest OLA titer was achieved (0.71 mg/L) (Fig. 2). However, FadD and MCS overexpression

208

only resulted in a slight increase to OLA compared with MCS alone. Previous studies showed

209

that different strains of E. coli show different production ability for different metabolites.35, 36 In

210

this regard, we tested the OLA production in BL21 (DE3) and E. coli K-12 MG1655 (DE3)

211

strain, and found that MG1655 (DE3) gives similar OLA titers compared with BL21 (DE3) (Fig.

212

S3). Specifically, MG-OLS-OAC-FadD-MCS produced ~0.8 mg/L OLA (Fig. S3), which is

213

comparable to that in BL-OLS-OAC-FadD-MCS (0.71 mg/L). Overall, while these results

214

demonstrate the importance of auxiliary enzymes for increasing malonyl-CoA and hexanoyl-CoA

215

supply, the low OLA titers mandated us to investigate alternative approaches for further

216

improving precursor supply.

217

Integration of the synthetic olivetolic acid pathway with a β-oxidation reversal for

218

precursor supply. In contrast to the above approach which relied on native metabolite pools or

219

exogenous acid addition for malonyl-CoA and hexanoyl-CoA supply, an alternative to further

220

improve OLA production involves the integrated engineering of pathways leading to precursor

221

synthesis. With malonyl-CoA generated directly from acetyl-CoA, the availability of this

222

intermediate may play a critical role in OLA production. Furthermore, the role of acetyl-CoA

223

becomes even more important when considering potential routes for generating hexanoyl-CoA.

224

In prior studies, both fatty acid biosynthesis (FAB) and β-oxidation reversal (r-BOX) pathways

225

have been employed for the production of hexanoic acid.24, 37, 38 However, the FAB pathway

10 ACS Paragon Plus Environment

Page 10 of 30

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

226

operates with acyl carrier protein (ACP) intermediates that are directly converted to carboxylic

227

acid products through the expression of heterologous specific short-chain C6-ACP thioesterase.38,

228

39

229

such as FadD, is required. Furthermore, the FAB pathway also requires malonyl-CoA as the

230

extender unit during elongation,40 resulting in increased competition for malonyl-CoA. In

231

contrast, r-BOX operates with CoA intermediates, utilizes acetyl-CoA as extender unit and can

232

directly generate hexanoyl-CoA. With this pathway initiating from acetyl-CoA and requiring an

233

additional 2 acetyl-CoA molecules to generate hexanoyl-CoA, ensuring high intracellular levels

234

of this acetyl-CoA intermediate are critical.

For conversion of hexanoic acid to hexanoyl-CoA, expression of fatty acyl-CoA synthetase,

235

To this end, we sought to exploit an engineered strain (JST10 (DE3)24) which has been

236

previously utilized for hexanoic acid synthesis through r-BOX. In addition to containing

237

chromosomal expression constructs for the required thiolase (BktB), β-ketoacyl-CoA reductase

238

(FadB), β-hydroxyacyl-CoA dehydratase (FadB), and trans-enoyl-CoA reductase (egTER)

239

r-BOX modules, this E. coli MG1655 derivative has fermentative product pathways (e.g. lactate,

240

succinate, acetate and ethanol) and thioesterases (e.g. tesA and tesB among others) deleted to

241

ensure adequate acetyl-CoA supply and minimize the loss of acyl-CoA intermediates. As such,

242

this strain is a promising background strain the expression of the synthetic OLA pathway (Fig.

243

3A). Furthermore, the increased acetyl-CoA supply in this strain may also provide a means of

244

utilizing ACC, opposed to MCS with exogenous malonate, for increasing malonyl-CoA

245

availability.

246

Integration of r-BOX and the OLA biosynthesis pathway in JST10 (DE3) expressing OLS

247

and OAC resulted in 2.8 mg/L OLA, nearly 30-fold higher than that in BL21 (DE3) (~0.1 mg/L),

248

even in the absence of hexanoate addition (Fig. 2). This result demonstrates the potential for

11 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

249

OLA production from a single carbon source (glycerol) through utilizing r-BOX for generating

250

hexanoyl-CoA. To evaluate if hexanoyl-CoA supply was still a limiting factor in this strain, we

251

also conducted experiments in which 4 mM hexanoate was supplied. Under these conditions,

252

JST10-OLS-OAC produced ~2.3-fold higher OLA (6.5 mg/L) indicating that improving

253

hexanoyl-CoA availability could potentially increasing OLA production. Interestingly, similar

254

increases to OLA titer upon hexanoate supplementation was observed both with and without

255

FadD expression under these conditions (Fig. 3B).

256 257 258

Figure 3. Production of olivetolic acid (OLA) in engineered E. coli JST10 (DE3). (A) In JST10 (DE3), a functional r-BOX was achieved by overexpression of BktB thiolase (TH), FadB hydroxyacyl-CoA

12 ACS Paragon Plus Environment

Page 12 of 30

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

259 260 261 262 263 264 265 266 267 268 269 270 271

dehydrogenase (HR), FadB enoyl-CoA hydratase (EH) and egTER enoyl-CoA reductase (ER). Fermentative by-product (lactate, succinate, ethanol, acetate) pathways were blocked through deletion of ldhA, frdA, adhE, pta, and poxB. OLS-OAC core enzymes and FadD-ACC auxiliary enzymes were recruited for OLA production. (B) Olivetolic acid titers in different strains. Engineered E. coli strains were grown in LB-like MOPS medium+2% (wt/v) glycerol supplemented with 4 mM hexanoate where indicated. For strains harboring MCS, 12 mM malonate sodium was also included. (C) Flaviolin biosynthesis pathway for measuring malonyl-CoA availability. Upper, flaviolin biosynthesis pathway: 5 malonyl-CoA are condensed by RppA to form flaviolin, which has a specific absorbance at wavelength of 340 nm. Bottom, RppA was expressed with different MCS/ACC auxiliary enzymes for characterization of malonyl-CoA availability in JST10 (DE3) strain. Engineered E. coli strains were cultured in LB-like MOPS medium+2% (wt/v) glycerol. For strains harboring MCS, 12 mM malonate sodium was also included.

272

overexpression of auxiliary enzymes. Although MCS was identified as the most effective

273

malonyl-CoA supply strategy for OLA production in resting cells biotransformation experiments

274

with BL21 (DE3) (Fig. 2), in actively growing cultures of JST10 (DE3) the overexpression of

275

ACC

276

JST10-OLS-OAC-ACC

277

JST10-OLS-OAC (2.8 mg/L) (p-value < 0.05). JST10-OLS-OAC-MCS (with 12 mM malonate

278

supplementation) produced 3.6 mg/L of olivetolic acid, which is a 28% increase compared to

279

JST10-OLS-OAC (p-value < 0.05) (Fig. 3B). This is likely caused by two factors, both leading

280

to higher levels of acetyl-CoA. First, the distinct metabolic backgrounds of BL21 (DE3) and

281

JST10 (DE3), as the deletion of acetyl-CoA competitive and consumption pathways in JST10

282

(DE3) is likely to result in increased availability of acetyl-CoA for malonyl-CoA generation.

283

Second, actively growing cultures of JST10 (DE3) should also lead to higher levels of

284

acetyl-CoA (and higher production of malonyl-CoA upon ACC overexpression). To confirm

285

increased malonyl-CoA supply in this background and conditions, a heterologous malonyl-CoA

286

availability indicator pathway was introduced. Flaviolin biosynthesis requires a PKS III catalysis

287

that uses malonyl-CoA as both primer and extender unit.41 Specifically, the polyketide synthase

288

RppA41 from Streptomyces griseus, which iteratively condenses 5 molecules of malonyl-CoA to

We then assessed the impact of malonyl-CoA supply on OLA production through the

resulted

in

the

highest

produced

increase 8.4

in

OLA

titer

mg/L of olivetolic

13 ACS Paragon Plus Environment

(Fig. acid,

3B).

Specifically,

3-fold

higher than

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

289

form flaviolin (which has a specific absorbance at the wavelength of 340 nm), was introduced

290

into JST10 (DE3) strains (Fig. 3C). While no significant increase in A340 was observed upon the

291

combined overexpression of RppA and MCS (with 12 mM malonate supplementation) compared

292

to RppA only, the overexpression of ACC with RppA lead to a significant increase in absorbance

293

(Fig. 3D) (p-value < 0.05). These results provide further evidence that ACC overexpression is an

294

effective strategy to increase malonyl-CoA supply in JST10 (DE3), which in turn leads to higher

295

OLA production.

296

Given the impact of auxiliary enzymes and individually increasing hexanoyl-CoA and

297

malonyl-CoA supply in JST10 (DE3), we next evaluated their combination in conjunction with

298

the synthetic OLA pathway. As seen in Figure 3B, additional hexanoate supplementation during

299

fermentation with JST10-OLS-OAC-ACC resulted in a 2-fold increase in OLA titer to 16.6 mg/L

300

(p-value < 0.05) (Fig. 3B). Moreover, despite the overexpression of FadD having a negligible

301

impact with hexanoate feeding and OLS/OAC overexpression, combined with ACC

302

overexpression FadD significantly improved OLA titer (26.2 mg/L) (Fig. 3B). This indicates the

303

importance of both hexanoyl-CoA and malonyl-CoA supply, as either can become the limiting

304

factor as intracellular supply of each is increased. We also evaluated the dosage effect of

305

hexanoate supplementation on OLA production and found that feeding 4 mM hexanoate

306

contributes to the highest titer of OLA (Fig. S4). As such, coordinated increase in the supply of

307

hexanoyl-CoA and malonyl-CoA is critical for producing OLA at high levels. While external

308

addition to hexanoate was required here to increase titers, we also demonstrate a new application

309

of r-BOX. Although prior studies of engineering of r-BOX primarily focused on production of

310

short-chain fatty acids or alcohols,

311

employed to supply the starting CoA primer for polyketide biosynthesis.

23, 24, 42

here we demonstrated that this pathway can also be

14 ACS Paragon Plus Environment

Page 14 of 30

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

312

Optimization of fermentation conditions for olivetolic acid production. Following the

313

establishment of the best combination of enzymatic components for OLA production, we

314

attempted to optimize the fermentation conditions for further titer improvement with strain

315

JST10-OLS-OAC-FadD-ACC. This included evaluation of the impact of various temperatures,

316

working volumes, and inducer concentrations on OLA production (Fig. 4). Results showed that

317

37 °C (26.8 mg/L) was the optimal temperature for OLA production with a significant decrease

318

in titer at both 30 °C (17.6 mg/L) and 22 °C (5.0 mg/L) (Fig. 4A). Based on the identified

319

optimal temperature (37 °C), we further studied the impact of working volume (WV, X mL in 25

320

mL flask, X/25 mL) as a means of altering aeration. We found that the engineered strain had the

321

highest OLA titer of 26.8 mg/L at a WV of 15/25 mL (Fig. 4B). In addition to temperature and

322

WV, we further investigated the impact of inducer concentrations (IPTG and cumate) on OLA

323

production. In the engineered strain, genes encoding OLS, OAC, FadD, and ACC enzymes are

324

expressed under the control of inducible T7 promoter, for which IPTG serves as inducer.

325

Excessive IPTG addition has been reported to be toxic to E. coli cells and will cause inclusion

326

body formation for excessive proteins biosynthesis,43 resulting in inhibition of enzymatic

327

activities and thus decreased product biosynthesis.44 To this end, optimization of IPTG dosage

328

for OLA production is desirable. Results showed that the JST10-OLS-OAC-FadD-ACC strain

329

still produced 3.1 mg/L of OLA without addition of IPTG, likely due to leaky expression under

330

the T7 promoter (Fig. 4C). Upon induction by IPTG, OLA production increased significantly and

331

a positive correlation was observed between IPTG dosage and OLA titer up to 100 µM. With 100

332

µM IPTG, the JST10-OLS-OAC-FadD-ACC strain produced 34.8 mg/L of OLA (Fig. 4C), which

333

is 30% higher the best titers achieved by using 50 µM IPTG. However, excessive dosage of IPTG

334

( > 100 µM) was found to decrease OLA production.

15 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

335

We also optimized the dosage effect of the inducer cumate, which activates the expression

336

of enzymes in r-BOX by binding with CymR repressor.45 Results showed that, the

337

JST10-OLS-OAC-FadD-ACC strain produced the highest level of OLA at 46.3 mg/L when

338

cumate was added at 10 µΜ (Fig. 4D). It is not surprising that the optimal dosage of cumate

339

inducer is relatively lower than optimal IPTG inducer (100 µΜ) since cumate-inducible r-BOX

340

genes, i.e. bktB, fadB, egTER, were integrated at the single site (atoB, fadB and fabI loci,

341

respectively)37 of chromosomal DNA of JST10 (DE3) instead of plasmids, and thus a small

342

dosage of cumate should be enough for switching on the expression of these genes. Similar to

343

IPTG, excessive cumate dosage also compromised OLA production (Fig. 4D).

344 345 346 347 348 349 350 351 352 353

Figure 4. Optimization of fermentation conditions for olivetolic acid production by JST10-OLS-OAC-FadD-ACC. The engineered E. coli strain was cultured in LB-like MOPS medium+2% (wt/v) glycerol in 25 mL shake flasks. (A) Effects of different temperature on OLA production with gene expression induced by 50 µM IPTG and 100 µM cumate. (B) Effects of different working volume (WV) on OLA production with gene expression induced by 50 µM IPTG and 100 µM cumate. (C) Effects of different IPTG dosages on OLA production with 100 µM cumate. (D) Effects of different cumate dosage on OLA production with 100 µM IPTG. For all experiments, inducers and 4 mM hexanoate were added after strains reached an OD550 ~0.4-0.8.

16 ACS Paragon Plus Environment

Page 16 of 30

Page 17 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

354

Olivetolic acid fermentation in bioreactor under controlled conditions. In order to obtain

355

higher OLA titer, a batch fermentation with precise parameter control was conducted using the

356

engineered strain JST10-OLS-OAC-FadD-ACC and the identified optimal fermentation

357

conditions. Under these conditions, cell growth of JST10-OLS-OAC-FadD-ACC reached the

358

highest OD550 of 8 (corresponding cell mass is approximately 2.64 g/L) at 48 hours. During the

359

first 24 hours, JST10-OLS-OAC-FadD-ACC consumed a total of ~8.5 g/L of glycerol, 79 mg/L

360

of hexanoate and produced approximately 80 mg/L of OLA (Fig. 5A). To our knowledge, this is

361

the highest OLA titer achieved in any wild type or engineered microorganism. In addition,

362

moderate accumulation of pyruvate was observed in the fermentation medium for removal of

363

downstream byproducts pathways (Fig. 3A). Also, although pta was deleted in JST10, a small

364

amount of acetate still produced, which might be due to presence of other unknown endogenous

365

transferase within E. coli (Fig. S5). Formation of both unwanted byproducts compromised the

366

OLA yield. Further, we continued to analyze the toxicity of OLA to E. coli, and revealed that up

367

to 100 mg/L of OLA did not impact the final cell mass of E. coli (Fig. S6), which excluded the

368

possibility that the product toxicity compromised OLA production.

369 370 371 372 373 374 375 376 377

Figure 5. Olivetolic acid (OLA) production and stability. (A) OLA fermentation in bioreactor with controlled conditions. Fermentation was performed in 400 mL MOPS medium with 30 g/L glycerol in 500 mL bioreactor (Infors). Cultures were grown at 37 ºC with an initial OD550 of 0.07, 100 µΜ IPTG, 10 µM cumate and 4 mM hexanoate were added when OD550 reached 0.4-0.8, the pH was maintained at 7.0 by using 1.5 M H2SO4 and 3 M NaOH, the dissolved oxygen level was also monitored. (B) Olivetolic acid stability assays in the absence/presence of E. coli MG1655 (DE3) cells. The initial olivetolic acid titer was ~110 mg/L. (C) Olivetol stability assay in the absence/presence of E. coli MG1655 (DE3) cells. All the olivetolic acid/olivetol stability assays were conducted in 400 mL MOPS medium with 30 g/L glycerol in 500 mL Infores bioreactor. In the

17 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

378 379 380

presence of E. coli cells, E. coli initial inoculum was set as OD550 ~0.07. OLA, olivetolic acid; OLO, olivetol.

381

Furthermore, we also observed that with the increase of fermentation time, OLA titer

382

decreased. Specifically, from 24 hours to 48 hours OLA concentration decreased by 50% to 37.4

383

mg/L. We hypothesize that the observed decrease in OLA concentration could be due to: 1) OLA

384

is inherently unstable and will degrade spontaneously in aqueous environments; 2) OLA can be

385

metabolized by E. coli cells. We further analyzed the stability of OLA in the absence and

386

presence of wild type E. coli cells (Fig. 5B) and determined that even in the absence of E. coli

387

MG1655 (DE3), OLA levels decreased over time, with the majority of OLA decarboxylated to

388

olivetol (Fig. 5B). In the presence of E. coli cells, although a portion was still observed to form

389

olivetol, OLA was degraded more than in the absence of cells, indicating at least a fraction of

390

OLA was metabolized by E. coli (Fig. 5B). Conversely, to further determine whether olivetol can

391

spontaneously convert to OLA or be metabolized by E. coli cells, we conducted a similar

392

experiments with olivetol. Results showed that olivetol can neither spontaneously convert to

393

OLA nor be metabolized by E. coli cells under our experimental conditions (Fig. 5B). Olivetol

394

seems more stable than OLA in aqueous environments, which might be related to the absence of

395

a carboxyl group.46

396 397

CONCLUSIONS

398

Despite the structural complexity of plant polyketides, the tractable starting units required

399

for their synthesis enables a synthetic approach for their production in which PKS and cyclase

400

components can be integrated with pathways for the generation of primer and extender units.

401

Here, functional expression and characterization of the Cannabis sativa OLA synthase and

402

cyclase enzymes confirmed their requirement for the synthesis of OLA from hexanoyl-CoA and 18 ACS Paragon Plus Environment

Page 18 of 30

Page 19 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

403

malonyl-CoA. Through the direct integration of OLS and OAC with modules of the β-oxidation

404

reversal aimed at generating hexanoyl-CoA, we demonstrate the synthesis of the plant natural

405

product OLA in engineered E. coli from a single carbon source. By further combining these

406

pathways with auxiliary enzymes for additional hexanoyl-CoA and malonyl-CoA generation, we

407

also identified the supply of these precursors as a key limiting factor in OLA synthesis. Through

408

combinational utilization of these auxiliary enzymes and optimization of fermentation conditions,

409

we achieved an OLA titer of ~ 80 mg/L. This represents the first report of OLA synthesis in E.

410

coli, and further demonstrates the potential for microbial cell factories to overcome the

411

limitations of direct plant extraction or chemical synthesis to produce plant-based natural

412

products.

413 414

METHODS

415

Strains and culture conditions. All strains used in this study are listed in Table 1. E. coli BL21

416

(DE3) and JST10 (DE3) were employed as the host strains. Luria-Bertani (LB) medium was

417

used for culturing E. coli cells for plasmid construction. Modified M9Y medium (6.7 g/L

418

Na2HPO4, 3 g/L KH2PO4, 0.5 g/L NaCl, 1 g/L of NH4Cl , 20 g/L glucose, 10 g/L yeast extract, 2

419

mM MgSO4, and 0.1 mM CaCl2) was used for the resting cells biotransformation experiments

420

with all BL21 (DE3) derivatives strains24, 42, 47. The “LB-like” MOPS medium used for JST10

421

(DE3) strains contains 125 mM MOPS, supplemented with 20 g/L glycerol (or 30 g/L in batch

422

fermentation and olivetolic acid/olivetol stability assays), 10 g/L tryptone, 5 g/L yeast extract, 5

423

mM calcium pantothenate, 2.78 mM Na2HPO4, 5 mM (NH4)2SO4, 30 mM NH4Cl, 5 µM sodium

424

selenite, 100 µM FeSO4.24 When necessary, ampicillin, spectinomycin and kanamycin were

425

added at final concentrations of 100, 50 and 50 mg/L, respectively.

426

Construction of plasmids. All oligonucleotide primers used in this study are listed in Table S1. 19 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

17, 18

Page 20 of 30

427

Codon optimized OLS and OAC

428

synthesized by GeneArt (Invitrogen) and then inserted into the first and second multiple cloning

429

site of pETDuet-1, respectively, resulting into pET-P1-OLS-P2-OAC. FadD and FadK were

430

PCR-amplified from E. coli K-12 MG1655 genomic DNA and inserted into the first cloning site

431

of pCDFDuet-1 to obtain pCDF-P1-FadD/FadK. The malonyl-CoA synthetase (MCS) gene from

432

Bradyrhizobium japonicum 27 was codon optimized and inserted into the second multiple cloning

433

site of pCDFDuet-1 to obtain pCDF-P2-MCS. AccA, AccB, AccC and AccD were

434

PCR-amplified from MG1655 genomic DNA with a ribosome binding site (RBS) (underlined,

435

Table S1) and assembled into the second multiple cloning site of pCDFDuet-1 through Gibson

436

Assembly Cloning Kit (NEB) to obtain pCDF-P2-ACC.

437

Table 1. Strains and plasmids used in this study. Plasmids/strains

from Cannabis sativa for expression in E. coli were

Genetic characteristics

Source

pBR322 ori with PT7; AmpR

Novagen

Plasmids pETDuet-1

R

pCDFDuet-1

CDF ori with PT7; Sm

Novagen

pET-P1-OLS

pETDuet-1 carrying ols

This study

pET-P1-OAC

pETDuet-1 carrying oac

This study

pET-P1-OLS-P2-OAC

pETDuet-1 carrying ols and oac

This study

pCDF-P1-FadD

pCDFDuet-1 carrying fadD

This study

pCDF-P1-FadK

pCDFDuet-1 carrying fadK

This study

pCDF-P2-MCS

pCDFDuet-1 carrying mcs

This study

pCDF-P2-ACC

pCDFDuet-1 carrying accA, accB, accC, accD

This study

pCDF-P1-FadD-P2-MCS

pCDFDuet-1 carrying fadD and mcs

This study

pCDF-P1-FadK-P2-MCS

pCDFDuet-1 carrying fadK and mcs

This study

pCDF-P1-FadD-P2-ACC

pCDFDuet-1 carrying fadD and accA, accB, accC, accD

This study

pCDF-P1-FadK-P2-ACC

pCDFDuet-1 carrying fadK and accA, accB, accC, accD

This study

pET-P1-RppA

pETDuet-1 carrying rppA

This study

Host strain for enzymes expression

Lab collection

E. coli Strains E. coli BL21 (DE3)

BL-OLS-OAC

∆frdA ∆ldhA ∆pta ∆adhE ∆poxB ∆yciA ∆ybgC ∆ydiI ∆tesA ∆fadM ∆tesB ∆fadE DE3 FRT-cymR-PCT5-fadB ∆fadA::zeo FRT-cymR-PCT5-bktB ∆atoB 24 FRT-cymR-PCT5-egTER at fabI chromosomal location BL21 (DE3) with pET-P1-OLS-P2-OAC This study

BL-OLS-OAC-FadD

BL21 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadD

This study

BL-OLS-OAC-FadK

BL21 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadK

This study

BL-OLS-OAC-MCS

BL21 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P2-MCS

This study

E. coli JST10 (DE3)

20 ACS Paragon Plus Environment

Page 21 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

BL-OLS-OAC-ACC

BL21 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-ACC

This study

BL-OLS-OAC-FadD-MCS

BL21 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadD-P2-MCS

This study

BL-OLS-OAC-FadK-MCS

BL21 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadK-P2-MCS

This study

BL-OLS-OAC-FadD-ACC

BL21 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadD-P2-ACC

This study

BL-OLS-OAC-FadK-ACC

BL21 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadK-P2-ACC

This study

MG-OLS-OAC

MG1655 (DE3) with pET-P1-OLS-P2-OAC

This study

MG-OLS-OAC-FadD-MCS

MG1655 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadD-P2-MCS

This study

MG-OLS-OAC-FadD-ACC

MG1655 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadD-P2-ACC

This study

JST10-OLS-OAC

JST10 (DE3) with pET-P1-OLS-P2-OAC

This study

JST10-OLS-OAC-FadD-MCS

JST10 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadD-P2-MCS

This study

JST10-OLS-OAC-FadD-ACC

JST10 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadD-P2-ACC

This study

JST10-OLS-OAC-FadD

JST10 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P1-FadD

This study

JST10-OLS-OAC-ACC

JST10 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P2-ACC

This study

JST10-OLS-OAC-MCS

JST10 (DE3) with pET-P1-OLS-P2-OAC and pCDF-P2-MCS

This study

JST10-RppA

JST10 (DE3) with pET-P1-RppA

This study

JST10-MCS-RppA

JST10 (DE3) with pET-P1-RppA and pCDF-P2-MCS

This study

JST10-ACC-RppA

JST10 (DE3) with pET-P1-RppA and pCDF-P2-ACC

This study

438 439

In vitro production of olivetolic acid. E. coli BL21 (DE3) was used for expression of

440

His-tagged OLS and OAC proteins, from their respective pET-P1-OLS and pET-P1-OAC

441

constructs. BL21 (DE3) strains containing His-tagged OAC or OLS genes were grown at 37 °C

442

in 0.5 L LB medium with ampicillin. Enzyme expression was induced by addition of

443

isopropyl-β-D-thiogalactoside (IPTG) to a final concentration of 0.4 mM, when OD550 of the

444

culture was between 0.4-0.8. After 18 h of induction at 37 °C, cells were harvested by

445

centrifugation at 12,000 rpm, 4 °C, 10 min. The cell pellet was resuspended in lysis buffer (20

446

mM Tris-HCl, 0.5 M NaCl, 5 mM imidazole, 0.1% Triton-X 100, pH 8.0) and subjected to

447

sonication using a Sonifier SFX250 (Branson). Following centrifugation (10,000 rpm, 4 °C, 30

448

min), the supernatant containing soluble protein fraction was recovered and filtered through a

449

0.45 µm filter. Recombinant His-tagged proteins were purified using TALON metal affinity resin

450

(Clontech). Soluble protein extract was applied to 1 ml packed column of the resin, and after

451

washing the unbound proteins with wash buffer (20 mM Tris-HCl, 0.5 M NaCl, pH 8.0)

452

supplemented with 20 mM imidazole, the His-tagged enzymes were eluted from the column with 21 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

453

elution buffer containing 250 mM imidazole. Purified His-tagged enzymes were concentrated to

454

a final concentration of 2 mg/mL and elution buffer was exchanged with storage buffer (12.5

455

mM Tris-HCl, 50 mM NaCl and 2 mM DTT) at 4 °C using Amicon ultrafiltration centrifugal

456

devices. The concentrated enzymes were stored at -80 °C for enzyme activity assays. Enzyme

457

assays were performed in a 500 µL total reaction volume containing 100 mM potassium

458

phosphate buffer (pH 7.0), 200 µM hexanoyl-CoA, 400 µM malonyl-CoA, 10 µg OLS, and 30

459

µg OAC (when included).18 The reaction mixture was incubated at 20 °C for 16 h and 20 µL

460

sulfuric acid (H2SO4) was added to terminate the reaction.

461

Resting cell biotransformations for olivetolic acid production. One milliliter (1 mL) of

462

overnight cultures of recombinant E. coli strains was inoculated in 50 mL fresh LB medium in

463

250 mL shake flask with ampicillin, and cultivated at 37 °C, 200 rpm. When OD550 reached

464

approximately 0.4-0.8, 0.5 mM IPTG was added. The cultures were then incubated at 22 °C for

465

15 h. Cells were then harvested by centrifugation, washed with fresh M9Y medium and

466

resuspended in 50 mL M9Y medium to OD550 ~ 3 and supplied with 4 mM hexanote for

467

biotransformation experiments.15, 48, 49 An additional 12 mM sodium malonate was added when

468

malonyl-CoA synthetase (MCS) was expressed for malonyl-CoA synthesis. Following incubation

469

at 22 °C for 48 h, the fermentation broth supernatants were extracted by equal volume of ethyl

470

acetate, evaporated by nitrogen and resuspended in 1 mL methanol for HPLC-MS analysis by

471

using Agilent 1200 HPLC system and Bruker MicroToF ESI LC-MS System. The column used

472

was Shim-pack XR-ODS II C18, 2.0 mm×75 mm (Shimadzu). HPLC conditions were as follows:

473

solvent A = 0.1% formic acid in H2O; solvent B = methanol; flow rate = 0.25 ml min-1; 0–2.5

474

min, 95% A and 5% B; 2.5–20 min, 95% A and 5% B to 5% A and 95% B; 20–23 min, 5% A and

475

95% B; 23–24 min, 5% A and 95% B to 95% A and 5% B; 24–30 min, 95% A and 5% B.15, 48, 49

22 ACS Paragon Plus Environment

Page 22 of 30

Page 23 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

476

Fermentation conditions for olivetolic acid production in shake flasks. Modified LB-like

477

MOPS medium using glycerol as carbon source was used for all fermentations. 24 Fermentations

478

were conducted in 25-mL Pyrex Erlenmeyer flasks (narrow mouth/heavy duty rim, Corning)

479

filled with 5-20 mL of the MOPS medium with 20 g/L glycerol and sealed with foam plugs

480

filling the necks. 24 A single colony of the desired strain was cultivated overnight (14-16 h) in LB

481

medium with appropriate antibiotics and used as the initial inoculum at the OD550 ~0.07. After

482

inoculation, flasks were incubated at 37 °C and 200 rpm until OD550 reached 0.4-0.8, at which

483

point IPTG (0-500 µM), cumate (0-500 µM) and hexanoate (4 mM) were added. 12 mM sodium

484

malonate was also added when malonyl-CoA synthetase (MCS) was expressed for malonyl-CoA

485

synthesis. Flasks were then incubated under the same conditions for 48 h post-induction unless

486

otherwise stated.

487

Olivetolic acid fermentation in bioreactor with precise parameter control. Fermentations

488

were performed in 400 mL MOPS medium with 30 g/L glycerol in a 500 mL bioreactor (Infors)

489

at 37 °C. An overnight seed culture was used to inoculate the bioreactor to an of OD550 ~0.07 and

490

when the OD550 reached 0.4-0.8, 100 µM IPTG, 10 µM cumate and 4 mM hexanoate were added.

491

pH was maintained at 7.0 by using 1.5 M sulfuric acid (H2SO4) as acid solution and 3 M

492

potassium hydroxide (NaOH) as base solution. The air flowrate was set at 50 mL/min, stirring

493

speed was set at 720 rpm. The dissolved oxygen (DO) level was set at 100% at the beginning,

494

with DO level monitored but not controlled during the whole fermentation period.

495

Olivetolic acid/olivetol stability analysis. For olivetolic acid stability assays, 500 bioreactors

496

with the above described media and conditions were utilized. The initial olivetolic acid

497

concentration was ~110 mg/L for olivetolic acid stability analysis. In the presence of E. coli cells,

498

MG1655 (DE3) was employed as the testing strain. An overnight seed culture was used to

23 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 30

499

inoculate the bioreactor an OD550 ~0.07. pH was maintained at 7.0 by using 1.5 M H2SO4 and 3

500

M NaOH. The air flowrate was set at 50 mL/min.

501

Similar operation was performed for olivetol stability assay, only changing the initial olivetolic

502

acid to olivetol (~80 mg/L).

503

GC-FID/MS analysis. Quantification of olivetolic acid was conducted via GC-FID analysis

504

using an Agilent 7890 B gas chromatograph equipped with an Agilent 5977 mass spectroscope

505

detector (Agilent) and an HP-5ms capillary column (0.25 mm internal diameter, 0.25 µm film

506

thickness, 30 m length; Agilent). Sample preparation was conducted as follows:

507

samples were transferred to 5 mL glass vials (Fisher Scientific), 4-pentylbenzoic acid (final

508

concentration 50 mg/L) was added as internal standard. Then 80 µL of H2SO4 and 340 µL of

509

30 % (wt/v) NaCl solution were added for pH and ionic strength adjustment. Two milliliters of

510

hexane was added for extraction. Vials were sealed with Teflon-lined septa (Fisher Scientific),

511

secured with caps, and rotated at 60 rpm for 2 h. The samples were then centrifuged for 2 min at

512

6500 rpm to separate the aqueous and organic layers. After centrifugation, 1.5 mL of the top

513

organic layer was transferred to new 5 mL glass vial and evaporated under a stream of nitrogen.

514

Then, 100 µL pyridine and 100 µL of N,O-Bis(trimethylsilyl)trifluoroacetamide (BSTFA) were

515

added to the dried extract for derivatization at 70 °C for 1 h. After cooling to room temperature,

516

200 µL of derivatization product was transferred to vials (Fisher Scientific) for GC-MS analysis

517

according to the following method: 1 µL were injected into the GC, which was run in splitless

518

mode using helium gas as a carrier gas with a flow rate of 1 mL/min. The injector temperature

519

was 280 °C and the oven temperature was initially held at 50 °C for 3 min and then raised to

520

250 °C at 10 °C/min and held for 3 min.

521

Statistical analysis. The two-tailed t-test method was employed to analyze the statistical

24 ACS Paragon Plus Environment

50

2 mL culture

Page 25 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

522

ACS Synthetic Biology

significance of all data in this study and p-value < 0.05 is deemed statistically significant.

523 524

ABBREVIATIONS

525

OLA, olivetolic acid; OLO, olivetol; PDAL, pentyl diacetic acid lactone; HTAL, hexanoyl

526

triacetic acid lactone; FadD, long chain fatty acid CoA-ligase; FadK, short chain acyl-CoA

527

synthetase; ACC, acetyl-CoA carboxylase; MCS, malonyl-CoA synthetase; r-BOX, reversal of

528

β-oxidation;

529

N,O-Bis(trimethylsilyl)trifluoroacetamide.

IPTG,

isopropyl-β-D-thiogalactopyranoside;

BSTFA,

530 531

SUPPORTING INFORMATION

532

Table. S1. Primers used in this study

533

Fig. S1. Expression of OLS and OAC in BL21 (DE3)

534

Fig. S2. Mass spectrometry results of olivetolic acid, olivetol and PDAL by using GC-MS

535

Fig. S3. Comparison of BL21 (DE3) with MG1655 (DE3) for olivetolic acid production.

536

Fig. S4. Effect of fed hexanoate dosage on olivetolic acid production

537

Fig. S5. Byproducts formation during olivetolic acid (OLA) production.

538

Fig. S6. Toxicity of olivetolic acid (OLA) to E. coli

539 540

AUTHOR INFORMATION

541

Corresponding Author

542

*E-mail: [email protected]

543

ORCID:

544

Ramon Gonzalez: 0000-0003-4797-6580

25 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

545

Zaigao Tan: 0000-0003-4528-0723

546

James M. Clomburg: 0000-0002-3133-9555

547

Author Contributions

548

R.G. designed research; Z.T. and J.M.C. performed research; Z.T. and J.M.C. analyzed data; Z.T.,

549

J.M.C. and R.G. wrote the paper.

550

Notes

551

R.G. owns shares of Bioactive Ingredients Corporation.

552 553

ACKNOWLEDGEMENTS

554

This work was supported by Bioactive Ingredients Corporation. The funders had no role in study

555

design, data collection and analysis, decision to publish, or preparation of the manuscript. The

556

authors thank Seokjung Cheong for assistance with gene cloning, Shivani Garg for help with

557

enzyme purification, and Seohyoung Kim for assistance with bioreactor fermentations.

26 ACS Paragon Plus Environment

Page 26 of 30

Page 27 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

REFERENCES 1. 2. 3. 4.

5.

6. 7. 8. 9.

10.

11. 12. 13. 14.

15. 16.

17.

18.

19. 20.

21. 22. 23.

Zhou, W., Zhuang, Y., Bai, Y., Bi, H., Liu, T., and Ma, Y. (2016) Biosynthesis of phlorisovalerophenone and 4-hydroxy-6-isobutyl-2-pyrone in Escherichia coli from glucose. Microb Cell Fact. 15, 149. Yu, D. Y., Xu, F. C., Zeng, J., and Zhan, J. X. (2012) Type III polyketide synthases in natural product biosynthesis. IUBMB life. 64, 285-295. Li, H., Liang, J., Chen, H., Ding, G., Ma, B., and He, N. (2016) Evolutionary and functional analysis of mulberry type III polyketide synthases. BMC Genomics. 17, 540. Ismed, F., Farhan, A., Bakhtiar, A., Zaini, F., Nugraha, Y. P., Putra, O. D., and Uckusa, H. (2016) Crystal structure of olivetolic acid: a natural product from Cetrelia sanguinea (Schaer.), Acta Crystallogr E. 72, 1587-1589. Costain, W. J., Tauskela, J. S., Rasquinha, I., Comas, T., Hewitt, M., Marleau, V., and Soo, E. C. (2016) Pharmacological characterization of emerging synthetic cannabinoids in HEK293T cells and hippocampal neurons. European journal of pharmacology. 786, 234-245. Tai, S., and Fantegrossi, W. E. (2014) Synthetic Cannabinoids: Pharmacology, Behavioral Effects, and Abuse Potential. Current addiction reports. 1, 129-136. Huestis, M. A. (2007) Human cannabinoid pharmacokinetics. Chemistry & biodiversity. 4, 1770-1804. Andre, C. M., Hausman, J. F., and Guerriero, G. (2016) Cannabis sativa: The Plant of the Thousand and One Molecules. Frontiers in plant science. 7. Hazekamp, A., Simons, R., Peltenburg-Looman, A., Sengers, M., van Zweden, R., and Verpoorte, R. (2004) Preparative isolation of cannabinoids from Cannabis sativa by centrifugal partition chromatography. J Liq Chromatogr R T. 27, 2421-2439. Yoshimatsu, K., Iida, O., Kitazawa, T., Sekine, T., Kojoma, M., Makino, Y., and Kiuchi, F. (2004) Growth characteristics of Cannabis sativa L. cultivated in a phytotron and in the field. Kokuritsu Iyakuhin Shokuhin Eisei Kenkyujo hokoku = Bulletin of National Institute of Health Sciences. 16-20. Yao, J., Weng, Y. Q., Dickey, A., and Wang, K. Y. (2015) Plants as Factories for Human Pharmaceuticals: Applications and Challenges. International Journal of Molecular Sciences. 16, 28549-28565. Chittiboyina, A., and Khan, I. (2012) Short synthesis of olivetolic acid via Directed ortho-Metalation (DoM). Planta Medica. 78, P_42. Liu, X., Ding, W., and Jiang, H. (2017) Engineering microbial cell factories for the production of plant natural products: from design principles to industrial-scale production. Microb Cell Fact. 16, 125. Ajikumar, P. K., Xiao, W. H., Tyo, K. E., Wang, Y., Simeon, F., Leonard, E., Mucha, O., Phon, T. H., Pfeifer, B., and Stephanopoulos, G. (2010) Isoprenoid pathway optimization for Taxol precursor overproduction in Escherichia coli. Science. 330, 70-74. Bai, Y., Yin, H., Bi, H., Zhuang, Y., Liu, T., and Ma, Y. (2016) De novo biosynthesis of Gastrodin in Escherichia coli. Metab Eng. 35, 138-147. Ro, D. K., Paradise, E. M., Ouellet, M., Fisher, K. J., Newman, K. L., Ndungu, J. M., Ho, K. A., Eachus, R. A., Ham, T. S., Kirby, J., Chang, M. C., Withers, S. T., Shiba, Y., Sarpong, R., and Keasling, J. D. (2006) Production of the antimalarial drug precursor artemisinic acid in engineered yeast. Nature. 440, 940-943. Taura, F., Tanaka, S., Taguchi, C., Fukamizu, T., Tanaka, H., Shoyama, Y., and Morimoto, S. (2009) Characterization of olivetol synthase, a polyketide synthase putatively involved in cannabinoid biosynthetic pathway. FEBS Lett. 583, 2061-2066. Gagne, S. J., Stout, J. M., Liu, E., Boubakir, Z., Clark, S. M., and Page, J. E. (2012) Identification of olivetolic acid cyclase from Cannabis sativa reveals a unique catalytic route to plant polyketides. Proc Natl Acad Sci U S A. 109, 12811-12816. Krivoruchko, A., Zhang, Y., Siewers, V., Chen, Y., and Nielsen, J. (2015) Microbial acetyl-CoA metabolism and metabolic engineering. Metab Eng. 28, 28-42. Vadali, R. V., Horton, C. E., Rudolph, F. B., Bennett, G. N., and San, K. Y. (2004) Production of isoamyl acetate in ackA-pta and/or ldh mutants of Escherichia coli with overexpression of yeast ATF2. Appl Microbiol Biotechnol. 63, 698-704. Lin, F., Chen, Y., Levine, R., Lee, K., Yuan, Y., and Lin, X. N. (2013) Improving fatty acid availability for bio-hydrocarbon production in Escherichia coli by metabolic engineering. PLoS One. 8, e78595. Zhou, S., Iverson, A. G., and Grayburn, W. S. (2008) Engineering a native homoethanol pathway in Escherichia coli B for ethanol production. Biotechnol Lett. 30, 335-342. Dellomonaco, C., Clomburg, J. M., Miller, E. N., and Gonzalez, R. (2011) Engineered reversal of the beta-oxidation cycle for the synthesis of fuels and chemicals. Nature. 476, 355-359.

27 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

24. 25.

26. 27. 28. 29. 30. 31.

32. 33. 34. 35.

36.

37.

38. 39.

40. 41. 42.

43. 44.

45.

46. 47.

Kim, S., Clomburg, J. M., and Gonzalez, R. (2015) Synthesis of medium-chain length (C6-C10) fuels and chemicals via beta-oxidation reversal in Escherichia coli. J Ind Microbiol Biotechnol. 42, 465-475. Yang, J., Xiong, Z. Q., Song, S. J., Wang, J. F., Lv, H. J., and Wang, Y. (2015) Improving heterologous polyketide production in Escherichia coli by transporter engineering. Appl Microbiol Biotechnol. 99, 8691-8700. Kim, Y. S., and Bang, S. K. (1985) Malonyl coenzyme A synthetase. Purification and properties, Journal of Biological Chemistry. 260, 5098-5104. Kim, Y. S., and Kang, S. W. (1994) Steady-state kinetics of malonyl-CoA synthetase from Bradyrhizobium japonicum and evidence for Malonyl-AMP formation in the reaction. Biochemical Journal. 297, 327-333. Wolfe, A. J. (2005) The acetate switch. Microbiology and molecular biology reviews : MMBR. 69, 12-50. Davis, M. S., Solbiati, J., and Cronan, J. E., Jr. (2000) Overproduction of acetyl-CoA carboxylase activity increases the rate of fatty acid biosynthesis in Escherichia coli. J Biol Chem. 275, 28593-28598. Shi, S. B., Chen, Y., Siewers, V., and Nielsen, J. (2014) Improving production of malonyl coenzyme A-derived metabolites by abolishing Snf1-dependent regulation of Acc1. mBio. 5, e01130-14.. Campbell, J. W., Morgan-Kiss, R. M., and Cronan, J. E. (2003) A new Escherichia coli metabolic competency: growth on fatty acids by a novel anaerobic beta-oxidation pathway. Molecular Microbiology. 47, 793-805. Morgan-Kiss, R. M., and Cronan, J. E. (2004) The Escherichia coli fadK (ydiD) gene encodes an anerobically regulated short chain acyl-CoA synthetase. J Biol Chem. 279, 37324-37333. Dirusso, C. C., and Black, P. N. (2004) Bacterial long chain fatty acid transport: gateway to a fatty acid-responsive signaling system. J Biol Chem. 279, 49563-49566. Ford, T. J., and Way, J. C. (2015) Enhancement of E. coli acyl-CoA synthetase FadD activity on medium chain fatty acids. PeerJ. 3, e1040. Jarboe, L. R., Zhang, X., Wang, X., Moore, J. C., Shanmugam, K. T., and Ingram, L. O. (2010) Metabolic engineering for production of biorenewable fuels and chemicals: contributions of synthetic biology. J Biomed Biotechnol. 2010, 761042. Monk, J. M., Charusanti, P., Aziz, R. K., Lerman, J. A., Premyodhin, N., Orth, J. D., Feist, A. M., and Palsson, B. O. (2013) Genome-scale metabolic reconstructions of multiple Escherichia coli strains highlight strain-specific adaptations to nutritional environments. Proc Natl Acad Sci U S A 110. 20338-20343. Clomburg, J. M., Blankschien, M. D., Vick, J. E., Chou, A., Kim, S., and Gonzalez, R. (2015) Integrated engineering of beta-oxidation reversal and omega-oxidation pathways for the synthesis of medium chain omega-functionalized carboxylic acids Metabolic Engineering. 28, 202-212. Torella, J. P., Ford, T. J., Kim, S. N., Chen, A. M., Way, J. C., and Silver, P. A. (2013) Tailored fatty acid synthesis via dynamic control of fatty acid elongation. Proc Natl Acad Sci U S A. 110, 11290-11295. Jing, F. Y., Cantu, D. C., Tvaruzkova, J., Chipman, J. P., Nikolau, B. J., Yandeau-Nelson, M. D., and Reilly, P. J. (2011) Phylogenetic and experimental characterization of an acyl-ACP thioesterase family reveals significant diversity in enzymatic specificity and activity. Bmc Biochemistry. 12. Tee, T. W., Chowdhury, A., Maranas, C. D., and Shanks, J. V. (2014) Systems metabolic engineering design: fatty acid production as an emerging case study. Biotechnology and Bioengineering. 111, 849-857. Funa, N., Ohnishi, Y., Ebizuka, Y., and Horinouchi, S. (2002) Properties and substrate specificity of RppA, a chalcone synthase-related polyketide synthase in Streptomyces griseus. J Biol Chem. 277, 4628-4635. Cheong, S., Clomburg, J. M., and Gonzalez, R. (2016) Energy- and carbon-efficient synthesis of functionalized small molecules in bacteria using non-decarboxylative Claisen condensation reactions. Nat Biotechnol. 34, 556-561. Baneyx, F. (1999) Recombinant protein expression in Escherichia coli. Curr Opin Biotechnol. 10, 411-421. Sriubolmas, N., Panbangred, W., Sriurairatana, S., and Meevootisom, V. (1997) Localization and characterization of inclusion bodies in recombinant Escherichia coli cells overproducing penicillin G acylase. Appl Microbiol Biotechnol. 47, 373-378. Mullick, A., Xu, Y., Warren, R., Koutroumanis, M., Guilbault, C., Broussau, S., Malenfant, F., Bourget, L., Lamoureux, L., Lo, R., Caron, A. W., Pilotte, A., and Massie, B. (2006) The cumate gene-switch: a system for regulated expression in mammalian cells. BMC Biotechnol. 6, 43. Schink, B., Philipp, B., and Muller, J. (2000) Anaerobic degradation of phenolic compounds. Die Naturwissenschaften. 87, 12-23. Neidhardt, F. C., Bloch, P. L., and Smith, D. F. (1974) Culture medium for enterobacteria. Journal of Bacteriology. 119, 736-747.

28 ACS Paragon Plus Environment

Page 28 of 30

Page 29 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

48. 49.

50.

Bai, Y., Bi, H., Zhuang, Y., Liu, C., Cai, T., Liu, X., Zhang, X., Liu, T., and Ma, Y. (2014) Production of salidroside in metabolically engineered Escherichia coli. Scientific reports. 4, 6640. Bi, H., Bai, Y., Cai, T., Zhuang, Y., Liang, X., Zhang, X., Liu, T., and Ma, Y. (2013) Engineered short branched-chain acyl-CoA synthesis in E. coli and acylation of chloramphenicol to branched-chain derivatives. Appl Microbiol Biotechnol. 97, 10339-10348. Clomburg, J. M., Vick, J. E., Blankschien, M. D., Rodriguez-Moya, M., and Gonzalez, R. (2012) A synthetic biology approach to engineer a functional reversal of the beta-oxidation cycle. ACS synthetic biology. 1, 541-554.

29 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Graphic Abstract 80x39mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 30 of 30